Mirati’s broadly acting tyrosine kinase inhibitor sitravatinib always looked like a long shot, so perhaps the most surprising thing about yesterday’s Sapphire trial failure is that it caught investors unawares: the group’s stock fell 13% this morning. The project looks dead despite an ongoing study in treatment-naive patients and Beigene-backed trials of tislelizumab combos. Much more important to Mirati, though, is the launch of Krazati in second-line NSCLC, expanding that product’s use into earlier lines and new indications, and the group’s remaining pipeline. This month, Krazati’s first full quarter of sales beat expectations, but there are already signs of a broad Kras flatlining. Expanding into the potentially lucrative first-line NSCLC market is therefore key for Mirati, but Keytruda-combo data here looked lacklustre; SVB analysts suggested that the addition of chemotherapy, being tested in Krystal-17, might help. Mirati has yet to nail down its strategy in first-line lung, and investors will hope that things move more quickly than in colorectal cancer, where a filing is now due by year-end. As for the pipeline, data on a PRMT5 inhibitor are up next, although this mechanism has disappointed in the past.

Two recent FDA approvals in amyotrophic lateral sclerosis bucked the usual trend of disappointments in the neurodegenerative disease. But it was back to business as usual this week, with two stumbles in quick succession from Wave Life Sciences and Apellis. In fact, a look at an Evaluate Vantage analysis of the late-stage ALS pipeline, carried out almost a year ago, shows that over half of these projects have since failed. Some developers are pressing on, with Prilenia and Clene still seeing a future for their assets. Such an optimistic stance is no doubt helped by the fact that Biogen and Ionis’s Qalsody (tofersen) got accelerated approval for Sod1-mutated ALS despite the antisense therapy failing its phase 3 trial, Valor. Perhaps the most optimistic of all is Brainstorm Cell Therapeutics, which after FDA knockbacks chose to file its project Nurown under protest, setting up an advisory committee meeting; a date has not yet been set. No such drama from Wave and Apellis, which have thrown in the towel on WVE-004 and Empaveli respectively. Notably one dose of WVE-004 performed statistically worse than placebo, echoing findings with Biogen and Ionis’s similarly-acting BIIB078 last year.

Mounjaro might be approaching the finish line in obesity, but Lilly is wasting no time in pushing forward its follow-on projects. Two new trials of its oral GLP-1 contender, orforglipron (LY3502970), are now listed on CT.gov, joining the open-label Achieve-4, which began recruiting in April. Only very top-line phase 2 data on orforglipron have been released so far, and Lilly has admitted that the agent is unlikely to promote as much weight loss as Mounjaro, which hits GIP as well as GLP-1. Analysts already expect big things, however: Evaluate Pharma’s sellside consensus sits at sales of $631m by 2028. Orforglipron is intended to target broader, primary care indications, the company said on its last earnings call, and perhaps meet demand in middle-income markets in China and other regions. Being “cheaper and easier” to make than injected Mounjaro also helps. Novo Nordisk's struggle to meet demand for its obesity drug, Wegovy, arguably creates a further incentive for obesity hopefuls to get trials of novel agents up and running quickly. Recruitment is presumably not going to be a problem, raising the prospect of very speedy development timelines for those coming behind. Expect the likes of Pfizer and Amgen to also move quickly here. 

What if the “skinny jab” were a pill? The oral version of Novo Nordisk’s obesity treatment Wegovy scored yesterday in its pivotal Oasis-1 study, and actually came in marginally better than its subcutaneous cousin. Novo did not go into great detail on safety, saying that the most common adverse events were gastrointestinal, with the vast majority being mild to moderate and consistent with the GLP-1 class. Filings will come this year, the group said; an oral version of the same compound, semaglutide, is already approved in diabetes as Rybelsus. Here, Novo is ahead of its arch-rival Lilly, which does not appear interested in developing an oral form of its highly impressive GIP/GLP-1 agonist Mounjaro. Novo is also ahead of Pfizer. Phase 2 data on that group’s oral GLP-1 danuglipron in diabetes, published yesterday, showed statistically significant improvements in blood sugar and body weight versus placebo at four months – but that pill must be taken twice daily, and it is not likely to be a serious competitor to oral Wegovy. Indeed the only cloud on Novo’s horizon is the major problem of ramping up sufficient production capacity.  

Regeneron’s lipid-lowering antibody Evkeeza has not been a big seller so far, but other groups are still hoping to emulate its mechanism, ANGPTL3 inhibition. And Arrowhead looks like it has got something just as good but more convenient: its RNA interference project ARO-ANG3 is given subcutaneously every quarter, versus Evkeeza’s once-monthly intravenous delivery. Caution is warranted when comparing across trials – especially as the latest data, presented today at the European Atherosclerosis Society Congress, come from an uncontrolled phase 2 study, called Gateway, in just 18 patients with homozygous familial hypercholesterolaemia (HoFH). ARO-ANG3 also looks better than Ionis’s anti-ANGPTL3 antisense asset vupanorsen, which Pfizer abandoned last year on disappointing LDL lowering. HoFH only affects around 1,300 US patients; a spokesperson for Arrowhead told Evaluate Vantage it is focused on this population for now, although it could go broader in future: it has also tested ARO-ANG3 in a broad mixed dyslipidaemia population in the phase 2 Arches-2 trial. Others targeting ANGPTL3 include Lilly, with the Dicerna-originated RNA interference project solbinsiran in phase 2, and Verve Therapeutics, which plans to start clinical trials of its in vivo base editing candidate VERVE-201 in the second half of 2024.

This story has been updated to include comments from Arrowhead on future development plans.

Novartis really does love lentiviral, but maybe not as much as some had hoped. The Swiss group today bought Avrobio’s cystinosis gene therapy candidate AVR-RD-04 for $87.5m, preferring this to acquiring Avrobio outright, as had been rumoured in February. A takeout would not have broken the bank. Earlier this year some sellside analysts suggested a $200m price, but Avrobio’s stock has come off since then and the group was worth a mere $35m on Friday. Perhaps Novartis was always only interested in AVR-RD-04, an autologous ex vivo modified haematopoietic stem cell project designed to increase cystinosin, which is defective in cystinosis. The asset is in an investigator-sponsored phase 1/2 trial, from which data on six patients were presented last week at the ASGCT meeting. Avrobio had planned to start its own trial this year; this will now become Novartis’s responsibility. For Avrobio, which only had enough cash to last until the first quarter of next year, this is a reprieve, since it will now be able to keep going until the end of 2024. Still, with the early nature of its pipeline, this only looks like a temporary solution unless the group can drum up interest in its other assets.

Buying Vectivbio, developer of a GLP-2 asset for short bowel syndrome, is a logical move for Ironwood, a group that has long focused on gastrointestinal disease. But the $1bn price represents a big bet on a crucial phase 3 readout, due by year end. Ironwood needs the project, apraglutide, to get to market swiftly as its main patents expire in 2030 – the asset has been around for some time. Apraglutide was discovered by Ferring, which first sold it on in 2012, and eventually landed in the hands of Vectivbio and its venture backers in 2020. Those VC backers, including Orbimed and Forbion, owned around 53% of Vectivbio at the end of 2022, having pumped in $145m over two rounds. While this might seem like a nice payday today’s sale, at $17 per share, comes in at the same price as Vectivbio’s 2021 IPO. The stock has been largely underwater since, so perhaps the investors will not be too unhappy. It is also notable that this takeover comes without any contingent payments, surprising given the looming data readout, and suggesting that this might have been a competitive process. Either way, Ironwood will soon know whether it has overpaid.

Another complete response letter for Intercept and its Nash project seems inevitable in the wake of Friday’s advisory committee drubbing. On the question of whether the benefits of Ocaliva outweigh its risks 12 panelists voted no and two yes, with two abstaining. On a call today Intercept executives held back from completely pulling the plug on further work in Nash ahead of the FDA’s final verdict, due by 22 June, but they made it pretty clear that this was on the cards. “We will immediately pivot to profitability” if accelerated approval is turned down, said the group's chief executive, Jerry Durso. This will presumably involve shutting down the large Regenerate trial, from which outcome data are still being collected, and unwinding any commercial investments already made. Some analysts project a move into the black for Intercept by 2025; with $337m in debt due by 2026, generating cash is already an important consideration. Executives attempted to turn attention to Ocaliva in its already approved indication, primary biliary cholangitis, but investors clearly have doubts about this planned pivot. Intercept stock sank 29% in morning trade, plumbing a record low.

FDA clearance of Beta Bionics’ version of an artificial pancreas could position the company as a takeover target. But who might buy, and how much might they pay? Speaking to Evaluate Vantage last year, executive chair Edward Damiano insisted that the iLet technology would appeal to a different population versus those using existing artificial pancreas systems like Medtronic’s recently approved MiniMed 780G, owing to the fact that the simpler MiniMed interface does not tend to allow the kind of granular insulin control artificial pancreas users like. So Medtronic or other players here like Abbott or Dexcom could be in the frame as buyers. So might makers of more traditional diabetes monitoring and therapeutic technologies like Ascensia or Lifescan, or even the biopharmas that manufacture insulin products – Beta Bionics counts Novo Nordisk, Lilly and Zealand Pharma among its venture backers. As for how much a notional buyer might pay, Beta Bionics last raised money in February 2022 at a valuation of just over $300m. With tuck-in deals increasingly in vogue in today’s cash-strapped environment, and Damiano saying he was open to the idea of a sale, Beta Bionics might not remain independent for long.

Johnson & Johnson investors face a nervous wait for results of the Mariposa study, and today they have something else to worry about. Astrazeneca’s Flaura2 trial has been toplined positive, suggesting – depending on the extent of the hit, of course – that even a convincing win in Mariposa might not count for much. That scenario already looked uncertain since Mariposa passed an interim analysis without being halted for efficacy, suggesting that any benefit might be smaller than hoped for. Flaura2 and Mariposa comprise front-line lung cancer patients with a specific EGFR mutation, a space effectively owned by Astra’s Tagrisso. In Mariposa J&J wants to beat Tagrisso with its Rybrevant plus lazertinib combo, but Flaura2 puts a fly in the ointment: the Astra trial adds chemo on top of Tagrisso and, since this has now been said to be statistically and clinically significant for PFS, Mariposa’s Tagrisso monotherapy comparator might not stay relevant for long. Much will depend on the size of the PFS benefit, which will no doubt prompt cross-trial comparisons to tease out any effect seen in Mariposa, and nothing is yet known about overall survival. Things will become clearer by the end of 2023.